Cost per kilometre hides where the money actually goes.
A blended cost per kilometre treats every drop the same. It is not. The expensive part of a delivery is not the distance, it is the stop: the loading, the waiting, the liftgate, the paperwork, the cash handling. Industry research shows last mile alone is 40 to 53 percent of total logistics cost. Time-driven costing puts a number on each of those minutes.
Cost and Profitability Consulting · 150+ models since 2010 · TDABC
Cost to serve in logistics varies 5 to 10 times between customers, because cost is driven by stops, handling and access, not by distance, and last mile alone is 40 to 53 percent of total logistics cost. Averaging cost per kilometre or per parcel hides this. TDABC assigns the real minutes of each delivery to the customer that caused them, so a low-volume, high-touch account stops being subsidised by an efficient one.
Same distance, opposite economics.
Cost per kilometre assumes the road is the cost. It is not. Loading, the number of stops, a liftgate, returns, paperwork and cash handling are all time, and time is cost. Two customers exactly the same distance from the depot can cost wildly different amounts to serve, and a blended rate charges them the same. The efficient, full-pallet account quietly pays for the awkward, multi-drop one, and nobody on the pricing side can see it happening.
SAME DISTANCE, DIFFERENT COST
Illustrative. Both customers are 40 km from the depot. The driving cost is identical; everything that makes Customer B expensive happens at the stop, where the average refuses to look.
Cost follows cause, minute by minute.
A delivery cost is built, not averaged. Base loading, a rate per kilometre, minutes per stop, and conditional terms that switch on only when the delivery demands them: a liftgate, extra paperwork, a cash payment. Each term carries the capacity cost rate of the resource it uses, so the cost lands on the customer that caused it.
Delivery cost = 15 min loading + 0.85 / km x distance + 8 min per stop + 25 min if liftgate required + 10 min paperwork & cash
Illustrative time equation. The conditional terms are where two same-distance deliveries diverge.
From a blended rate to a served-cost decision.
Cost each delivery
Run the time equation per drop, loaded with the real capacity cost rate of driver, vehicle and depot.
Roll up to the customer
Sum deliveries to the account. The 5 to 10 times spread between customers becomes visible.
Find the subsidy
See which efficient accounts are paying for which high-touch ones, and by how much.
Reprice or redesign
Adjust minimums, frequency or surcharge, or change the service, so price meets real cost.
What companies published this quarter
Evidence reviewed 6 August 2026. Each item below is quoted from the company’s own published statement and links to it. None of these companies is a client of Cost and Profitability Consulting, and nothing here is a comment on how any of them is run.
- Smurfit Westrock (Ireland, 29 July 2026): reported a Q2 adjusted EBITDA margin of 14.2 per cent under pressure from significantly higher freight and input costs, recovered by pricing rather than by cost structure. Source
- CTT, Correios de Portugal (Portugal, 28 July 2026): cut its 2026 recurring operating result guidance after operating expenses grew faster than revenue, with half-year profit down 41.6 per cent. Source
- Ebro Foods (Spain, 29 July 2026): reported pasta division revenue flat at EUR 346.3 million, down 0.2 per cent, in a half marked by an unfavourable dollar and increased transport costs into the United States. Source
- Carrefour España (Spain, 24 July 2026): reported EUR 490 million of group cost savings in the first six months, roughly half the EUR 1,000 million annual target, on Spanish turnover of EUR 5,726 million at a 3.3 per cent operating margin. Source
Freight arrives in the accounts as one line. It is not caused by one line. It is caused by drop sizes, delivery windows, order frequency and return rates that differ enormously between customers, and a group margin cannot tell you which of them you are subsidising. That is the question cost-to-serve exists to answer.
Frequently asked questions
- What is cost to serve in logistics?
- The true cost of serving a route, stop or customer once stop time, handling, access and paperwork are loaded, not a blended cost per kilometre. Cost to serve varies 5 to 10 times between customers because cost is driven by the stop, not the distance, and last mile alone is 40 to 53 percent of total logistics cost.
- Why does cost per kilometre mislead?
- It assumes distance drives cost. It does not. Two customers the same distance away can cost very different amounts once you load loading time, the number of stops, a liftgate, returns and cash handling. Averaging cost per kilometre or per parcel spreads those minutes evenly, so a low-volume, high-touch account is silently subsidised by an efficient one.
- How do you measure cost to serve per delivery?
- With a delivery time equation: base minutes plus a rate per kilometre plus minutes per stop, plus conditional terms for liftgate, paperwork and cash. Multiply each by the capacity cost rate of the resource it consumes, and the cost lands on the customer that caused it rather than on the network average. The generic version of the method is set out in how to calculate cost to serve.
Find the delivery cost the average is hiding.
The Profit Check needs no data upload. It points to where your cost to serve and your pricing are most likely out of line, and what closing the gap is worth.
- Duration
- 12 to 15 minutes
- You receive
- Score, 7 dimensions, sector benchmark
- Price
- Free, no email needed
Who you would be talking to
Miguel Guimarães, Founding Partner
Cost and profitability practitioner for 25+ years. Lectured alongside Professor Robert S. Kaplan at the CFO conference in Amsterdam (2009).
Call +351 910 313 731
Workshops20-21 Oct · Online, ZoomReserve a seat
Reviewed by
Miguel Guimarães
Founding Partner, Cost and Profitability Consulting
More than 150 Time-Driven ABC engagements across 11 sectors since 2010, working within the Kaplan and Anderson framework.
About the author →Published Updated
Done. Advertising and measurement is now declined on this browser, so the LinkedIn Insight Tag does not load and nothing about your visit is shared with LinkedIn. You can turn it back on here at any time.
We are asking again. Your last answer is more than 12 months old, so it has been cleared. Nothing is switched on until you answer.
This notice has changed. Your previous answer was given to different wording, so it has been cleared and we are asking again. Nothing is switched on until you answer.
Session recording. Microsoft Clarity records how this page is used - mouse movement, clicks and scrolling - so we can improve it. Form fields are always hidden from the recording.
Advertising and measurement. The LinkedIn Insight Tag tells LinkedIn that you visited this site, so we can measure our advertising and reach people like you. This shares data with LinkedIn, a third party outside this site.
GPC. Your browser sends a Global Privacy Control signal, so the advertising purpose is declined automatically and we do not ask you about it.
- 1Time Driven Activity Based Costing
- 2Cost-to-Serve Analysis
- 3The Whale Curve
- 4TDABC vs ABC
- 5Make-or-Buy and Relevant Costs
- 6Cost-Volume-Profit (CVP) and Break-Even Analysis
- 7Customer Profitability Analysis
- 8Methods & Frameworks: how we cost, defensibly
- 9TDABC for Financial Services
- 10How to calculate cost to serve, step by step